NeMo-Aligner: Scalable Toolkit for Efficient Model Alignment

Gerald Shen, Zhilin Wang, Olivier Delalleau, Jiaqi Zeng, Yi Dong, Daniel Egert, Shengyang Sun, Jimmy J. Zhang, Sahil Jain, Ali Taghibakhshi, Markel Sanz Ausin, Ashwath Aithal, Oleksii Kuchaiev

Research Area: Alignment, Engineering for large LMs

Keywords: Model Alignment, RLHF, LLM, Helpfulness, Toolkit, Scalable System

TL;DR: Toolkit for aligning LLMs with techniques like RLHF, DPO, SteerLM and SPIN that can efficiently scale to over a thousand GPUs

Abstract:

Aligning Large Language Models (LLMs) with human values and preferences is essential for making them helpful and safe.
However, building efficient tools to perform alignment can be challenging, especially for the largest and most competent LLMs which often contain tens or hundreds of billions of parameters.
We create NeMo-Aligner, a toolkit for model alignment that can efficiently scale to a thousand GPUs for training the largest open-source LLMs such as Nemotron 4 340B and Llama 3.1 405B.
NeMo-Aligner comes with highly optimized and scalable implementations for major paradigms of model alignment such as: Reinforcement Learning from Human Feedback (RLHF), Direct Preference Optimization (DPO), SteerLM, and Self-Play Fine-Tuning (SPIN).
Additionally, our toolkit supports running most of the alignment techniques in a Parameter Efficient Fine-Tuning (PEFT) setting.
NeMo-Aligner is designed for extensibility, allowing support for other alignment techniques with minimal effort.
It is open-sourced with Apache 2.0 License and we invite community contributions at https://github.com/NVIDIA/NeMo-Aligner.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the COLM Code of Ethics on https://colmweb.org/CoE.html

Author Guide: I certify that this submission complies with the submission instructions as described on https://colmweb.org/AuthorGuide.html

Submission Number: 922